Understanding the SIP Protocol

From VoIPmonitor.org
Revision as of 16:49, 8 January 2026 by Admin (talk | contribs) (Rewrite: condensed verbose sections, preserved all tables/diagrams, reduced from 1157 to 663 lines)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Quick Navigation
Architecture & Basics Messages & Call Flow Advanced Topics

Core Components

Addressing & Transport

SIP Methods

Response Codes

Call Flows

Core Concepts

Registration

Extensions

Reference

Session Initiation Protocol (SIP) is an application-layer signaling protocol for creating, modifying, and terminating multimedia sessions over IP. SIP handles signaling; actual media uses RTP. Messages are text-based (like HTTP) using a request/response model with session descriptions in SDP.

Core specification: RFC 3261 (2002). Extended by numerous RFCs for reliability, events, IM, security, etc.

SIP Architecture

User Agent (UA)

An endpoint (softphone, IP phone) that initiates or receives calls. Has two roles:

  • UAC (User Agent Client) - initiates requests (caller)
  • UAS (User Agent Server) - responds to requests (callee)

Roles flip depending on direction - BYE sender becomes UAC for that transaction.

Proxy Server

Intermediary that routes SIP requests/responses on behalf of UAs. Performs routing logic, policy enforcement, authentication. Can be:

  • Stateful - maintains transaction state, can fork requests to multiple destinations
  • Stateless - simply forwards and forgets

Adds Via and optionally Record-Route headers before forwarding.

Registrar

Handles REGISTER requests - stores user's current location (IP/port) in the Location Service database. Maps Address-of-Record (AOR) like sip:alice@atlanta.com to Contact address of device.

Back-to-Back User Agent (B2BUA)

Acts as UAS to caller and UAC to callee, terminating one dialog and creating another. Common in PBX and SBC systems. Unlike proxies, maintains full call state and can modify signaling/media.

Addressing

URI Scheme Description Example
sip: Standard SIP URI sip:alice@atlanta.com
sips: Secure SIP (TLS required) sips:alice@atlanta.com
tel: Telephone number (RFC 3966) tel:+1-555-123-4567

SIP uses DNS SRV/NAPTR records (RFC 3263) to locate servers for a domain.

Transport

Transport Port Notes
UDP 5060 Most common, requires retransmission handling
TCP 5060 For large messages, reliable delivery
TLS 5061 Encrypted signaling
WebSocket 80/443 For web applications (RFC 7118)

SIP Messages

Two types: Requests (methods) and Responses (status codes). Format: start line, headers, blank line, optional body (usually SDP).

Core SIP Methods

Method Purpose Creates Dialog?
INVITE Initiate session/call (carries SDP offer) Yes
ACK Confirm receipt of INVITE final response No
BYE Terminate established session No (ends dialog)
CANCEL Cancel pending INVITE before answer No
REGISTER Register contact with server No
OPTIONS Query capabilities (ping/keepalive) No

Extension Methods

Method RFC Purpose
PRACK 3262 Acknowledge reliable provisional responses
SUBSCRIBE 3265 Subscribe to event notifications
NOTIFY 3265 Send event notification
PUBLISH 3903 Publish event state (presence)
INFO 2976 Mid-session info (DTMF)
REFER 3515 Request call transfer
MESSAGE 3428 Instant message
UPDATE 3311 Modify session in early dialog

SIP Response Codes

Class Meaning Common Codes
1xx Provisional 100 Trying, 180 Ringing, 183 Session Progress
2xx Success 200 OK, 202 Accepted
3xx Redirection 301 Moved Permanently, 302 Moved Temporarily
4xx Client Error 400 Bad Request, 401 Unauthorized, 404 Not Found, 486 Busy, 487 Request Terminated
5xx Server Error 500 Internal Error, 503 Service Unavailable
6xx Global Failure 603 Decline, 604 Does Not Exist Anywhere

Only final responses (2xx-6xx) terminate a transaction. Provisional (1xx) are informative only.

SIP Headers

Header Purpose Example
Via Route responses back; branch ID for matching Via: SIP/2.0/UDP host;branch=z9hG4bK...
From Caller identity + tag From: "Alice" <sip:alice@atlanta.com>;tag=1234
To Callee identity + tag (added in response) To: "Bob" <sip:bob@biloxi.com>;tag=5678
Call-ID Unique call identifier Call-ID: a84b4c76e66710@host
CSeq Request sequence number + method CSeq: 314159 INVITE
Contact Direct reachable address Contact: <sip:alice@192.0.2.4:5060>
Max-Forwards Hop limit (default 70) Max-Forwards: 70
Content-Type Body MIME type Content-Type: application/sdp
Record-Route Proxy stays in signaling path Record-Route: <sip:proxy;lr>
Authorization Authentication credentials Authorization: Digest username="alice"...

Dialog Identification

A SIP dialog is uniquely identified by: Call-ID + From-tag + To-tag

Example: INVITE with SDP

INVITE sip:bob@biloxi.com SIP/2.0
Via: SIP/2.0/UDP pc33.atlanta.com;branch=z9hG4bK776asdhds
Max-Forwards: 70
To: Bob <sip:bob@biloxi.com>
From: Alice <sip:alice@atlanta.com>;tag=1928301774
Call-ID: a84b4c76e66710@pc33.atlanta.com
CSeq: 314159 INVITE
Contact: <sip:alice@pc33.atlanta.com>
Content-Type: application/sdp
Content-Length: 142

v=0
o=alice 2890844526 2890844526 IN IP4 pc33.atlanta.com
s=Session SDP
c=IN IP4 pc33.atlanta.com
t=0 0
m=audio 49172 RTP/AVP 0
a=rtpmap:0 PCMU/8000

SDP Quick Reference

Line Meaning
v=0 SDP version (always 0)
o= Origin: username, session-id, version, address
c= Connection: where to send media (IP)
m=audio 49172 RTP/AVP 0 Media: type, port, protocol, payload types
a=rtpmap:0 PCMU/8000 Attribute: payload 0 = G.711 μ-law
a=sendrecv Direction: bidirectional

SIP Call Flow Example

Step Message Purpose
1 INVITE Initiate call (with SDP offer)
2 100 Trying Stop retransmissions (proxy-generated)
3 180 Ringing Callee alerting
4 200 OK Call answered (with SDP answer)
5 ACK Confirm 200 OK receipt
6 RTP Media session
7 BYE End call
8 200 OK Confirm termination

Canceling a Call

CANCEL uses same Call-ID and CSeq number as the INVITE. Only effective before final response - ignored if 200 OK already sent.

Transactions, Dialogs, and Sessions

Concept Scope Lifetime Identified By
Transaction Single request/response Seconds Branch ID, CSeq, Method
Dialog UA-to-UA relationship Minutes-hours Call-ID + From-tag + To-tag
Session Media exchange Duration of call SDP negotiation

ACK Behavior

For 2xx Response For non-2xx Response
Part of INVITE transaction? No (separate) Yes (same)
UAS behavior if no ACK Retransmits 200 OK Retransmits error response
Routing End-to-end via Route set Hop-by-hop

Registration

Action Expires Value Result
Initial registration e.g., 3600 Binding created
Refresh registration e.g., 3600 Binding updated
Unregister 0 Binding removed

Registration must be refreshed before expiration. Multiple devices can register same AOR - proxies fork INVITEs to all.

SIP Extensions

Reliability of Provisional Responses (RFC 3262)

Basic SIP sends 1xx responses unreliably. RFC 3262 adds PRACK method to acknowledge provisional responses. UAS includes Require: 100rel and RSeq header; UAC responds with PRACK.

Session Timer (RFC 4028)

Keep-alive mechanism for dialogs. Refresher sends periodic re-INVITE/UPDATE. If refresh fails, call terminates. Prevents zombie calls when one side crashes. Uses Session-Expires and Min-SE headers.

REFER Method (RFC 3515)

Used for call transfer. Alice sends Bob REFER with Refer-To: carol@domain. Bob initiates new INVITE to Carol. Transfer result reported via NOTIFY.

NAT Traversal

Extension RFC Purpose
rport 3581 Symmetric response routing (send to source IP:port, not Via address)
outbound 5626 Persistent flows, keep-alives for NAT bindings

Quick Reference: Extensions

Extension RFC Key Headers/Methods
100rel 3262 PRACK, RSeq, RAck
timer 4028 Session-Expires, Min-SE
replaces 3891 Replaces header
outbound 5626 Instance-ID, reg-id
gruu 5627 Contact with gr parameter

Monitoring and Troubleshooting

What to Monitor

Layer What to Track Why
Signaling INVITE/BYE timing, response codes, retransmissions Identifies call setup failures, routing issues
Media (RTP) Packet loss, jitter, latency, MOS Actual voice quality
Registration REGISTER success/failure Endpoint reachability

Common SIP Issues

Issue Symptoms Cause
One-way audio Caller hears, callee doesn't (or vice versa) NAT: SDP contains private IP
Call setup delays Long wait before ringing DNS problems, proxy timeouts
No ringback tone Silence while waiting for answer 180→183 conversion without RTP
Auth loops Repeated 401/407 responses Credential mismatch, realm error
Codec failure 488 Not Acceptable No common codecs

Ringback Tone Issue (180 vs 183)

Response Behavior Audio Source
180 Ringing Callee alerting Caller generates local ringback
183 Session Progress Early media Network sends RTP audio

Problem: Device converts 180→183 but doesn't send RTP. Caller expects network audio but receives silence.

Diagnosis:

  1. Check SIP History for 183 sent to caller
  2. Check RTP tab for packets after 183
  3. If RTP present but no ringback: caller device issue
  4. If no RTP: intermediate device issue - fix SBC/PBX config

Quality Metrics

Metric Good Acceptable Poor
MOS > 4.0 3.5 - 4.0 < 3.5
Packet Loss < 1% 1% - 3% > 3%
Jitter < 20ms 20-50ms > 50ms
Latency < 150ms 150-300ms > 300ms

Key RFCs

RFC Title Category
3261 SIP: Session Initiation Protocol Core
3262 Reliability of Provisional Responses (PRACK) Reliability
3263 Locating SIP Servers DNS/Routing
3264 An Offer/Answer Model with SDP Media
3265 SIP-Specific Event Notification Events
3515 The REFER Method Transfer
3581 Symmetric Response Routing (rport) NAT
4028 Session Timers in SIP Keep-alive
5626 Client-Initiated Connections (Outbound) NAT

See Also

AI Summary for RAG

Summary: Session Initiation Protocol (SIP) is an application-layer signaling protocol for establishing, modifying, and terminating multimedia sessions. Architecture includes User Agents (UAC/UAS), Proxy Servers (stateful/stateless routing), Registrars (store user locations), and B2BUAs (full call control). Core methods: INVITE (initiate), ACK (confirm), BYE (terminate), CANCEL (abort), REGISTER (location binding), OPTIONS (capabilities). Response classes: 1xx provisional, 2xx success, 3xx redirect, 4xx client error, 5xx server error, 6xx global failure. Key headers: Via (routing), From/To (identities with tags), Call-ID (unique identifier), CSeq (sequence), Contact (direct address), Record-Route (keep proxy in path). Dialog identified by Call-ID + From-tag + To-tag. Extensions include PRACK (reliable provisionals), session timers (keep-alive), REFER (transfer), rport/outbound (NAT traversal).

Keywords: SIP protocol, INVITE, ACK, BYE, CANCEL, REGISTER, OPTIONS, 180 Ringing, 183 Session Progress, early media, ringback tone, PRACK, dialog, session, transaction, SDP, Call-ID, CSeq, Via, From, To, Contact, Record-Route, rport, outbound, NAT traversal, session timer, REFER, transfer, proxy, registrar, B2BUA, response codes, 200 OK, 401 Unauthorized, 486 Busy, 487 Request Terminated

Key Questions:

  • How does SIP establish a call between two endpoints?
  • What is the difference between 180 Ringing and 183 Session Progress?
  • Why does a caller hear silence instead of ringback tone?
  • How do SIP headers like Via, Contact, and Record-Route work?
  • What is a SIP dialog and how is it identified?
  • What is the difference between transactions, dialogs, and sessions?
  • How does SIP registration work?
  • What are common SIP troubleshooting issues (one-way audio, auth failures)?
  • What SIP extensions exist for NAT traversal?
Retrieved from "https://www.voipmonitor.org/doc/index.php?title=Understanding_the_SIP_Protocol&oldid=10165"